\begin{table}[h!]
\centering
\small
\caption{Expenditures by Type}
\label{tab:het1}
\begin{tabularx} {\linewidth} {@{} l Z Z Z Z}
\toprule
\toprule
 & Wages & Services & Fuel & Inputs \\
\cmidrule{2-5}
 & (1) & (2) & (3) & (4) \\
\midrule
\textbf{Panel A: Training} & & & & \\
Treatment Effect & -0.390 & -0.598 & -0.289** & -0.105  \\
 & (0.285) & (0.394) & (0.132) & (0.189) \\
 &  &  &  &  \\
N & 405 & 405 & 405 & 405 \\
\midrule
\textbf{Panel B: Technical Assistance} & & & & \\
Treatment Effect & 0.161 & 0.592* & 0.429** & 0.682**  \\
 & (0.305) & (0.334) & (0.190) & (0.265) \\
 &  &  &  &  \\
N & 462 & 462 & 462 & 462 \\
\bottomrule
\bottomrule
\end{tabularx}
\begin{tablenotes}
\small
\begin{spacing}{1.0}
\item Notes: {Each row reports the results for a different outcome. Column 1 reports the coefficient of a bivariate regression of each outcome on the dummy indicating that the farm was offered technical assistance. Column 2 adds the outcome itself measured in the baseline as a control. Column 3 adds the pre-determined characteristics as additional controls. The sample is restricted to farmers who completed the training. Robust standard errors are reported in parentheses. *** p<0.01; ** p<0.05; * p<0.10}
\end{spacing}
\end{tablenotes}
\end{table}%
